Cyclic guanosine monophosphate (hereinafter, referred to as cGMP) functioning as a second messenger in cells is known to play an important role in various physiological functions including learning and memory behaviors.
On the postsynaptic site of the brain neural circuits, nitrogen monoxide (hereinafter, referred to as NO) biosynthesized by a nitrogen monoxide synthetase activates a guanylate cyclase, which is a cGMP synthetase. The activated guanylate cyclase biosynthesizes cGMP from guanosine triphosphate. The cGMP activates a cGMP-dependent protein kinase (hereinafter, referred to as PKG) to phosphorylate various proteins participating in synapse plasticity. The activation of the NO/cGMP/PKG cascade is known to participate in the induction of synapse plasticity (Long Term Potentiation; hereinafter, referred to as LTP) of the hippocampus known as a neural substrate for learning and memory behaviors (for example, see Non Patent Literature 1). A medicine activating the signal transmission of the cascade is known to improve LTP of the hippocampus and the learning behavior of animals, while a medicine inhibiting the cascade is known to exhibit the opposite action (Non Patent Literature 2). Therefore, from these findings, an increase in cGMP in the brain is anticipated to lead to an improvement of learning and memory behaviors.
cGMP is metabolized to 5′-GMP having no PKG activation action by a phosphodiesterase (hereinafter, referred to as PDE). The PDE is known to have 11 families, and PDE9 is known to metabolize specifically cGMP, and to be expressed in the brain, the spleen, the small intestine and the like (for example, see Non Patent Literature 3). That is, inhibition of PDE9 is anticipated to increase cGMP in brains. It is reported that a PDE9 inhibitor actually enhances hippocampus LTP, and improves the learning and memory behaviors in a novel-object recognition test/passive avoidance learning test or the like in animals (Non Patent Literature 4). Clinically, guanylate cyclase activity decreases and possibility of a decrease in the cGMP level is indicated in the superior temporal cortex of Alzheimer's disease patients, (Non Patent Literature 5). Therefore, the PDE9 has a possibility of having many close relations with pathologies of neurodegenerative diseases and psychiatric diseases, particularly with pathologies of cognitive dysfunctions and the like in the Alzheimer's disease, such as Alexander's disease, Alpers' disease, Alzheimer's disease, amyotrophic lateral sclerosis (ALS; known as Lou Gehrig's disease or motor neuron disease), ataxia-telangiectasia, Batten's disease (known also as Spielmeyer-Vogt-Sjogren-Batten's disease), Binswanger's dementia (subcortical angiosclerotic encephalopathy), bipolar disorder, bovine spongiform encephalopathy (BSE), Canavan's disease, chemotherapy induction dementia, Cockayne's syndrome, corticobasal degeneration, Creutzfeldt-Jakob's disease, depression, Down's syndrome, frontotemporal lobe degeneration (including frontotemporal dementia, semantic dementia and progressive nonfluent aphasia), Gerstmann-Straussler-Scheinker's disease, glaucoma, Huntington's disease (chorea), HIV related dementia, hyperkinesis, Kennedy's disease, Korsakoff s syndrome (amnesic confabulation syndrome), Krabbe's disease, Lewy-bodies dementia, progressive logopenic aphasia, Machado-Joseph's disease (spinocerebellar ataxia type 3), multiple sclerosis, multiple atrophy (olivopontocerebellar atrophy), myasthenia gravis, Parkinson's disease, Pelizaeus-Merzbacher's disease, Pick's disease, dementia presenilis (slight cognitive impairment), primary lateral sclerosis, primary progressive aphasia, radiation-induced dementia, Refsum's disease (phytanic acid storage disease), Sandhoffs disease, Schilder's disease, schizophrenia, semantic dementia, senile dementia, Shy-Drager syndrome, spinocerebellar ataxia, spinal muscle atrophy, Steele-Richardson-Olszewski's disease (progressive supranuclear palsy), and vascular amyloidosis and vascular dementia (multiple infarct dementia).
Recently, the following compound has been known which has PDE9 inhibitory activity and has a purpose of prevention or therapy of Alzheimer's disease (Patent Literature 1).

The above compound is a pyrazolopyrimidine derivative, and a compound having a structure totally different from a pyrazoloquinoline skeleton.
On the other hand, as a compound having a pyrazoloquinoline skeleton, the following compound described in Patent Literature 2 is known.
wherein a ring A is a benzene ring or the like; and R6 is a direct bond or the like.
However, a ring B in the above compound denotes a benzene ring or the like. Although it is stated that the above compound has inhibitory activity against PDE4 and is used for various types of inflammatory diseases, there is no description for implication of the inhibitory activity against PDE9, and the like.
As compounds having PDE9 inhibitory activity, the following compounds described in Patent Literature 3 and Patent Literature 4 are known.

Any of the above compounds is a quinoxaline derivative, and is a compound having a structure totally different from a pyrazoloquinoline skeleton.
As a compound having a pyrazoloquinoline skeleton and having PDE9 inhibitory activity, the following compound described in Patent Literature 5 is known.
wherein either R1 or R2 is a group represent by the formula

The structure of the above compound is restricted in R1 and R2, thus the compound is a compound having a structure totally different from the compound of the present invention.